Effects of moderate-intensity aerobic exercise on serum BDNF and motor learning in the upper-limb in patients after chronic-stroke: A randomized, controlled feasibility study with embedded health economic evaluation

Article type: Research Article

Authors: Maguire, Clare^{a; b; *} | Betschart, Martina^{a; b; e} | Pohl, Johannes^c | Primani, Francesca^{a; b} | Taeymans, Jan^{b; d} | Hund-Georgiadis, Margret^a

Affiliations: [a] REHAB, Clinic for Neurorehabilitation and Paraplegiology, Basel, Switzerland | [b] Physiotherapy, Department of Health, Bern University of Applied Science, Bern, Switzerland | [c] Department of Neurology, University Hospital Zurich, Zurich, Switzerland | [d] Faculty of Movement and Rehabilitation Sciences, Vrije Universiteit Brussel, Brussels, Belgium | [e] Kantonal Hospital Winterthur, Winterthur, Switzerland

Correspondence: [*] Address for correspondence: Clare C. Maguire, REHAB, Clinic for Neurorehabilitation and Paraplegiology, Im Burgfelderhof 40, 4055 Basel, Switzerland. E-mails: [email protected] and [email protected].

Abstract: BACKGROUND:Brain-derived neurotrophic factor (BDNF) promotes activity-dependent neuroplasticity and is released following aerobic-exercise. OBJECTIVE:Feasibility and efficacy of 1.Moderate-Intensity Cycle-Ergometer-Training (MI-ET) and 2.Low-Intensity Circuit-Training (LI-CT) on BDNF-serum-concentration in chronic-stroke and consequently efficacy of motor-learning in varying BDNF-concentrations (neuroplasticity being the substrate for motor-learning) via upper-limb robotic-training (RT) in both groups. METHODS:Randomised-control feasibility-study. 12-week, 3x/week intervention, 17 chronic-stroke-survivors randomized into: (1) MI-ET&RT or (2) LI-CT&RT. Both groups completed 40 mins MI-ET or LI-CT followed by 40 mins RT. Feasibility outcomes: (1) screening and enrollment-rates, (2) retention-rates, (3) adherence: (i) attendance-rates, (ii) training-duration, (4) adverse events. Primary clinical outcomes: 1. serum-BDNF changes pre-post training (immediate) and pre-training basal-levels over 12-weeks (long-term). 2.upper-limb performance with Action-Research-Arm-Test (ARAT). Additionally, feasibility of an embedded health economic evaluation (HEE) to evaluate health-costs and cost-effectiveness. Outcomes: cost-questionnaire return-rates, cost-of-illness (COI) and Health-Utitility-Index (HUI). RESULTS:21.5% of eligible and contactable enrolled. 10 randomized to MI-ET and 7 to LI-CT. 85% of training-sessions were completed in MI-ET (306/360) and 76.3% in LI-CT-group (165/216). 12-weeks: Drop-outs MI-ET-10%, LI-CT-43%. Clinical outcomes: No significant changes in immediate or long-term serum-BDNF in either group. Moderate-intensity aerobic-training did not increase serum-BDNF post-stroke. Individual but no group clinically-relevant changes in ARAT-scores. HEE outcomes at 12-weeks: 100% cost-questionnaires returned. Group-costs baseline and after treatment, consistently favouring MI-ET group. COI: (1-year-time-frame): MI-ET 67382 SD (43107) Swiss-Francs and LI-CT 95701(29473) Swiss-Francs. CONCLUSION:The study is feasible with modifications. Future studies should compare high-intensity versus moderate-intensity aerobic-exercise combined with higher dosage arm-training.

Keywords: Stroke, aerobic training, BDNF, circuit training, upper-limb robotic training, health economic evaluation

DOI: 10.3233/NRE-220239

Journal: NeuroRehabilitation, vol. 52, no. 3, pp. 485-506, 2023